Boron Production

The following methods have been proposed for the isolation of boron -

1. Reduction of boron sesqui-oxide with potassium, sodium, magnesium, or aluminium. According to Wohler and Deville, 10 parts of boron sesqui- oxide are mixed with 6 parts of sodium, covered with 5 parts of sodium chloride, and heated to redness. The magnesium process is discussed later,
2. Electrolysis of fused boron sesqui-oxide or borax.
3. Reduction of boron trifluoride by means of heated potassium.
4. Reduction of potassium borofluoride by means of heated potassium or magnesium.
5. Reduction of borax by heated magnesium or red phosphorus.
6. Reduction of boron trichloride by means of hydrogen.
7. Reduction of the oxide, bromide, or nitride of boron by heating with metallic calcium.
8. Dissociation of magnesium boride, "boron suboxide," and boron nitride in vacuo at high temperatures (1500° and above).

Pure boron was first obtained in 1909, by Weintraub. Previous to this, Moissan claimed to have prepared boron containing not more than 2 per cent, of impurities. According to Moissan, the early processes described by Guy-Lussac and Thenard, Berzelius, and Wohler and Deville, lead to products which, after purification by heating with water and hydrochloric acid, only contain 40 to 70 per cent, of boron.

According to the method adopted in its preparation, pure boron may be obtained either in small grains having a sub-metallic appearance or in the massive form, the lumps exhibiting a conchoidal fracture. It is very probable, but not certain, that in each of these forms boron is crystalline. The so-called crystalline varieties of boron described by the earlier chemists are, however, borides of aluminium.

Moissan, who considered that he had prepared nearly pure boron, described it as an amorphous, brown powder, and most of the earlier investigators described it in similar terms. The researches of Weintraub and of Ray have shown, however, that boron obtained by Moissan's method contains several per cent, of oxygen and that under certain conditions the oxygen content may reach as much as 16 per cent. In fact, the reduction of excess of either boron sesqui-oxide or borax with a suitable metallic reducing agent always leads to the production of an amorphous, brown powder containing oxygen. Weintraub refers to such a product as boron suboxide; it is most probably a solid solution of an oxide of boron, possibly B₄O₃, in boron. This so-called amorphous boron is rapidly converted into boric acid when gently warmed with 40 per cent, nitric acid. Under such treatment pure boron is hardly attacked. The small proportion of pure boron usually present in the amorphous product can therefore be readily isolated.

The practical methods for obtaining pure, or nearly pure boron, are as follows

The fluoride, generated in a lead or platinum apparatus to exclude silicon fluoride, and free from hydrogen fluoride, is led over sodium heated in a hard glass tube to the lowest temperature at which reduction can be effected. The product is extracted with water and hydrochloric acid in a platinum dish, when boron is left as a brown powder.

A more convenient process consists in substituting potassium borofluoride for boron trifluoride. Excess of the potassium salt is mixed with sodium in a crucible lined with magnesia, and heated to redness in a rapid stream of dry hydrogen. The boron is isolated as before.

Early experimenters used for this purpose 2 parts of oxide to 1 of metal, the reaction being considered to be as follows -

2B₂O₃ + 3Mg = Mg₃(BO₃)₂ + 2B.

Moissan increased the amount of oxide to 3 parts, thereby keeping the "amorphous boron" produced suspended in a very fluid mass of boron sesqui-oxide and away from the air. His method is as follows -

Boron sesqui-oxide (210 grams), free from sodium, calcium, and silicon compounds and recently fused and powdered, is mixed in an earthenware crucible with magnesium turnings (70 grams) free from iron and silicon. The crucible is covered and placed in a furnace already at a bright red heat. Within five minutes, a violent reaction takes place. The mass is heated for another ten minutes, then cooled and removed from the crucible. Externally the reaction product is black; internally it has a maroon colour, traversed in all directions by white crystals of magnesium borate. The black outer layers are removed and rejected. The residue is powdered and boiled with a large quantity of dilute hydrochloric acid. It is then heated with pure hydrochloric acid at the boiling-point of the latter for two hours, a procedure that is repeated six times. The solid residue is washed with water, heated with boiling 10 per cent, alcoholic potash, and again washed with water. Finally it is heated in a platinum dish for four hours with 50 per cent, hydrofluoric acid, washed with water, and dried, first on a porous plate and then in a vacuum over phosphoric anhydride. The product contains 88 to 90 per cent, boron and 2 to 4 per cent, magnesium (Weintraub), the preceding treatment having served to free it almost completely from boron sesqui-oxide, magnesium borate, boron nitride, and silica. To eliminate the remaining magnesium, the boron is mixed intimately with fifty times its weight of boron sesqui-oxide, introduced between two layers of the oxide in a large earthen crucible and heated to bright redness. The product, when cold, is treated by the process already described. In this manner it is possible to obtain a brown, amorphous powder containing only 0.4 per cent, of magnesium. It is advantageous, during each heating in the furnace, to enclose the crucible in a larger one brasqued with a finely powdered mixture of rutile and carbon. An alternative method of eliminating nitrogen is to carry out the preparation in an iron vessel lined with magnesia, a rapid stream of hydrogen being passed through it.

Mercury arc furnace for melting boron.

In the coarse of his investigations of the magnesium process, Weintraub found that if a higher temperature than that used by Moissan is employed, there is no difficulty in obtaining a product free from magnesium. To obtain the higher temperature, Moissan's procedure may be used, with charges of at least one kilogram of material, or the reaction may be effected in a graphite crucible heated to 1700° in a vacuum electric furnace. The high temperature product, however, contains not 4 to 5 per cent, but 14 to 16 per cent, of oxygen, and re-fusion with a large Excess of boron sesqui-oxide is without action upon it. If the ratio oxide metal is diminished below 2:1, the product obtained, after the usual chemical treatment, consists mainly of magnesium boride.

Magnesium boride, "boron suboxide," and boron nitride all dissociate when heated in vacuo to sufficiently high temperatures. With the boride, the change is perceptible at 1200°, and is rapid at 1500°. The suboxide and nitride, however, dissociate with more difficulty, particularly the latter. Owing to this dissociation, it is comparatively easy to obtain pure boron from the so-called " amorphous boron " or " boron suboxide," and at the same time to melt the boron. (i.) One method consists in placing the powder in a water-cooled copper cup which forms one of two copper electrodes, and passing an arc between them in vacuo or in hydrogen at a few millimetres pressure. Several pounds of boron may be prepared and melted at one operation in this manner, starting with boron suboxide (Weintraub). (ii.) Another method consists in using a " mercury arc furnace." In this case, Moissan's boron or boron suboxide should be compressed into rods and purified as far as possible by heating A boron rod D, mounted in a carbon holder to 1200° to 1500° in vacuo. C to which a lead B is attached, is placed within the cylindrical glass apparatus A (fig.). The apparatus may be filled with hydrogen or evacuated by means of suitable connections. The apparatus being evacuated, mercury is run in through the funnel E until the end of D is just covered. Hydrogen is then admitted up to 2.5 to 15 cms. pressure, the electrodes H and K connected to a suitable source of current, and mercury run into G until the end of D is clear of the mercury in the lower part F of the apparatus. An arc is thus. formed between the boron and the mercury in F, and, since at the surface of the latter great heat is developed, the end of D melts and falls off. As the rod D thus shortens, more mercury is run in from time to time from E. The bottomless alundum pot L prevents the arc from straying to the glass. In this manner, extremely pure boron may be obtained (Weintraub). (iii.) A third method, somewhat analogous to the preceding method (ii.), consists in focussing the cathode rays on the powdered material, an operation that must obviously be carried out in a nearly perfect vacuum.

The preceding methods for obtaining pure boron necessitate the employment of special forms of apparatus. A fourth method is very simple to carry out, but does not give the product in the massive form, (iv.) The " suboxide" is mixed with excess of magnesium or sodium and heated to bright redness in a stream of hydrogen; in the former case magnesium boride is formed and subsequently decomposed at the high temperature employed. The product is extracted successively with water, hydrochloric acid, and warm 40 per cent, nitric acid.

This method, also due to Weintraub, is based upon the reversible reaction

2BCl₃ + 3H₂ ⇔ 2B + 6HCl,

which has been studied by Pring and Fielding, and also by Besson and Fournier. A high temperature is necessary for the production of boron by this method, and is obtained by the use of an electric arc. The preparation is effected by running one or more alternating-current arcs between water- or air-cooled copper electrodes in a mixture of boron trichloride vapour and a large excess of hydrogen; a glass or copper containing vessel may be used. The gases are mixed by passing dry hydrogen over the surface of boron trichloride contained in a vessel cooled by a freezing mixture, or by allowing the chloride to drop into a vessel (kept a little above the room temperature) through which hydrogen is passed. The boron is partly thrown off as a fine powder and partly settles on the electrodes, where it grows into small rods. These eventually melt down to beads and fall off, and the process repeats itself. The boron powder, after washing with water, contains 99 to 99.5 per cent, of boron; the fused lumps are pure boron. By any of the processes already mentioned the boron powder may be purified and melted.